The specific aims of this study are to test the health behavior and acculturation hypotheses, which have been proposed to explain the paradox. These hypotheses posit that: (1) Latinos have more favorable cancer-related health behaviors and risk factor profiles than non-Latino whites, and (2) Health behaviors and risk factors become more unfavorable with greater acculturation. Specific health behaviors and risk factors to be studied are: smoking, alcohol use, exercise, and body mass index. Preventive health behaviors to be examined include: having received a Pap smear (yes/no) and clinical breast examination (CBE) (yes/no), both over the past 12 months; and a mammogram over the past 2 years (yes/no); and a routine physical examination. Hypotheses will be tested using data from the National Health Interview Survey (NHIS). Length of U.S. residence will be used as a proxy measure of acculturation. Many of the existing published NHIS reports do not present data on health behaviors by Latino groups, nor are appropriate controls made for factors such as age, SES, and access to health care. The proposed study will control for age and SES, and conduct detailed analyses of acculturation by Latino groups, as well as examine the effects of access to health care on various health behaviors. This study will, therefore, contribute to a systematic program of research attempting to account for the epidemiologic paradox of Latino mortality.